 Ann Dozier, RN, PhD (PI) › Community and Preventive Medicine; University of Rochester  Cindy R. Howard, MD, MPH › Pediatrics; Rochester General Hospital  Ruth A. Lawrence, MD › Pediatrics; University of Rochester  Cynthia K. Childs, MFA, MPH › Community and Preventive Medicine ;University of Rochester  Study conducted 2004 – 2006  Funded through APTR/CDC Cooperative Agreement › T 1267

 Effective outcome evaluation requires that › program components are delineated › the overall program(s) clearly depicted  Role of process evaluation ›Document what program participants received or were ‘exposed’ to ›Thereby avoid assumptions about what the program actually entailed

 This study evaluated breastfeeding outcomes among women who delivered at two hospitals with different breastfeeding programs

 Participating hospitals: › Baby Friendly designated (BFHI) › Community Hospital (CH) with well established breastfeeding program › The impact/outcome evaluation focused on  breastfeeding duration and exclusivity at 6 months  over 400 BF women enrolled from each hospital post-delivery/before discharge

Baby Friendly’s 10 steps used for inter-hospital program comparisons 1: Have a written breastfeeding policy that is routinely communicated to all health care staff. 2: Train all health care staff in skills necessary to implement this policy. 3: Inform all pregnant women about the benefits and management of breastfeeding. 4: Help mother initiate breastfeeding within an hour of birth. 5: Show mothers how to breastfeed and how to maintain lactation, even if they should be separated from their infants. 6: Give newborn infants no food or drink other than breastmilk, unless medically indicated 7: Practice rooming in– allow mothers and infants to remain together == 24 hours a day. 8: Encourage breastfeeding on demand. 9: Give no artificial teats or pacifiers to breastfeeding infants. 10: Foster the establishment of breastfeeding support groups and refer mothers to them on discharge from the facility.

 Three perspectives were compared across each of the 10 steps  hospital administrators  inpatient nurses  post partum mothers  The process evaluation was completed prior to analysis of the outcome data  This prevented findings about hospital outcomes to influence the interpretation the process evaluation data.

 Hospital administrators › 5 interviews with nursing/medical administrators at each hospital › Employed previously tested tool (Kovach, 1996); structured in-person interview › Assessment included all 10 steps › Answers were tallied according to Kovach's methodology for each step.  Even though the community hospital was not Baby Friendly designated, for comparison purposes it was assessed on the same steps/components.

 Inpatient nurses › Using previously developed survey (Lawrence)  with additional questions adapted from Kovach instrument that aligned with the Baby Friendly 10 steps › Full and part time nurses (BFH= 61; CH=43)  working on inpatient/mother-baby unit completed survey  anonymously, mailed to their home address › Nurses scored 9 of the 10 Baby Friendly steps (except whether the mothers had been informed prenatally about the benefits of breastfeeding (Step 3))

 Post partum mothers › Enrolled mothers completed an in-person interview about their background, their infant feeding decisions, support, previous experience and other factors thought to influence feeding decisions and duration. › At two weeks post discharge they completed a mailed survey that included questions extracted from the PRAMS survey that seek to capture mothers' hospital breastfeeding experience; (BFH= 341; CH=321)  Mothers' input could only be obtained reliably for six of the ten Baby Friendly Steps (e.g. they are not able to comment on presence of written policies or training).

 For comparisons, scores were standardized › ratings above 90% were considered fully implemented › partial implementation >75% › between group differences of <20% indicated rating concurrence › Six components were rated by all three groups  help initiating rooming in  showing how no pacifier  only breastmilk fostering support

STEP 6: Give newborn infants no food or drink other then breastmilk, unless medically indicated.  Administrators: Do staff have a clear understanding of what the few acceptable reasons are for prescribing food or drink other than breast milk for breastfeeding babies? › Response options: Yes/No  Nurses: What are healthy breastfed babies given as a "first feeding"? – Breastmilk; › Response options: Always/Most of the time (75-100%), Sometimes, Seldom/Not at all (25% or less)  Mothers: Was your baby fed only breast milk during your hospital stay? › Response options: Yes/No

STEP 6: Give newborn infants no food or drink other than breastmilk, unless medically indicated. Hospital Ratings:  Administrators (Yes): › BFH: 3 out of 3(100%) CH: 2.88 out of 3 (96%)  Nurses (Most): › BFH: 100% CH: 100%  Mothers (Yes): › BFH :70% CH: 56%

 For this step, both administrators and nurses at each hospital concurred that the program did promote that breastmilk only be given to breastfed infants. › Mothers from both hospitals indicated other wise (did not concur)  These findings imply that program implementation may not be as consistent as the internal (administrators/nurses) assessments indicate.

Hospital Administrators Scores Baby Friendly StepBFHCH #1 Written BF Policy #2 Train all health care staff #3 Inform all pregnant women33 #4 Help mother initiate BF #5 Show mothers how to BF32.92 #6 Give infants no food or drink other than breastmilk32.88 #7 Practice ‘rooming in’ #8 Encourage unrestricted BF33 #9 Give no pacifiers or artificial nipples33 #10 Foster establishment of BF support groups Average Score

 Based on the six common to administrators, nurses and postpartum mothers › administrators' rated  5 as Fully Implemented  1 as Partially Implemented › Nurses concurred with all administrator ratings. › Among the 5 Fully Implemented components, mothers' ratings concurred with 3 --help initiating --showing how --no pacifier

 Of the six chosen for comparison, › Administrators' rated  4 as Fully Implemented  1 as Partially Implemented › Nurses' ratings differed  3 as Fully Implemented  1 as Partially Implemented  They concurred with administrators on Full Implementation of 2 components › Mothers ratings  Of above 2 Fully Implemented, mothers' ratings concurred with 1  --no pacifier

 These hospitals had….  some similarities based on program descriptions AND  their overall administrator scores were similar, BUT….. › they differed substantively on extent of implementation

 This finding reinforces the importance of documenting implementation (through process evaluation) in order to correctly interpret and attribute outcome findings to a particular programmatic intervention.  Not only will this help explain between-hospital differences it will help better define those components that may contribute to outcome differences.